﻿This TwitterFFL_MFG_20190521_readme.txt file was generated on 20190521 by Lisa Jordan.

-------------------
GENERAL INFORMATION
-------------------

Title of Dataset: FFL Manufacturers Twitter Assessment, May 2019  

Author Information (Name, Institution, Address, Email)

  Lisa Jordan, Drew University, 36 Madison Ave. Madison, NJ, ljordan@drew.edu
  James Kalin, Drew University, 36 Madison Ave. Madison, NJ
  Colleen Dabrowski, Drew University, 36 Madison Ave. Madison, NJ

Date of data collection (single date, range, approximate date): 20190515
Geographic location of data collection: Madison, NJ

Information about funding sources or sponsorship that supported the collection of the data:
Digitial Humanities Summer Institute (DHSI), funded by Andrew W. Mellon grant, awarded to Drew University.

--------------------------
SHARING/ACCESS INFORMATION
-------------------------- 

Licenses/restrictions placed on the data, or limitations of reuse: as is

Recommended citation for the data:

Jordan, L., Kalin, J., Dabrowski, C. 2019. FFL Manufacturers Twitter Assessment, May 2019.  

Links/relationships to ancillary or related data sets: Original Twitter Analytics (Date(GMT), Handle, Name, Text, URL, Platform, Type, Retweet count, Favorite count) downloaded from Twitonomy [http://twitonomy.com/].

--------------------
DATA & FILE OVERVIEW
--------------------

File list (filenames, directory structure (for zipped files) and brief description of all data files):

TwitterFFL_MFG_20190523.xls; TwitterFFL_MFG_20190521.csv

TwitterFFL_MFG_20190521.xls contains 928 tweets, each coded for an additional attributes, and three ID codes.

Selections of the AggregateMFG in the first sheet are included in subsequent sheets, listing tweets for rifles, handguns, shotguns, most retweeted, most recent, and top 100 retweeted.

If data was derived from another source, list source:

If there are there multiple versions of the dataset, list the file updated, when and why update was made:

Original Twitter Analytics (Date(GMT), Handle, Name, Text, URL, Platform, Type, Retweet count, Favorite count) downloaded from Twitonomy [http://twitonomy.com/].

--------------------------
METHODOLOGICAL INFORMATION
--------------------------

Description of methods used for collection/generation of data: <Include links or references to publications or other documentation containing experimental design or protocols used in data collection>

First, major arms manufacturers were identified using the Annual Firearms Manufacturing and Export Report, 2017." Originally published by the Bureau of Alcohol, Firearms and Tobacco (ATF), 30 January 2019.  These data were converted from a .pdf to .csv file in order to total the number of guns produced per manufacturer, by location.  This dataset is available on Harvard Dataverse:

ATF. 2019. "Annual Firearms Manufacturing and Export Report, 2017." Originally published by the Bureau of Alcohol, Firearms and Tobacco (ATF), 30 January 2019, Washington, DC. Downloaded 14 May 2019 from https://www.atf.gov/file/133476/download as afmer_2017_final.pdf.  Data Compiled and Aggregated to .csv format by L. Jordan, J. Kalin, J. Jordan. 14 May. Available online from Harvard Dataverse.

For this project, we chose to explore the top manufacturers, which produced over 50 thousand firearms.  Several manufacturers operated at multiple locations.  Given that a significant portion of firearms enter the US as imports, and are not included in the manufacturing list, we identified top manufacturers in other countries with sizable exports of firearms to the US: Croatia (HS Produkt), Turkey (MKE, also Zenith Firearms), the Czech Republic (CZ firearms), and the Philippines (Armscor).  Within the list of major producers, a few did not have Twitter handles (Sturm, Ruger & Company, Colt’s Manufacturing Company, LLC, and HS Produkt).  We were able to identify twitter handles for 24 manufacturers, representing 68 percent of the domestic firearms production.  From these, we obtained the approximately 2,800 most recent tweets, ending May 15, 2019, from Twitonomy.  For coding purposes, from the recent tweets obtained, we explored the twenty most recent tweets, ending May 15, 2019, and the twenty most retweeted tweets.

Methods for processing the data: 

From the Twitonomy spreadsheets downloaded for each twitter handle.  ID codes were used to identify the twenty most recent tweets (excluding replies), and the top twenty tweets retweeted for each company handle.  In some cases there was an overlap among the most recent and most retweeted.  Then, the LJ, JK and CD coded characterized each tweet based on 19 characteristics, defined below: Handgun, Shotgun, Rifle, Attributes, Protection, Hunting, Recreation, 2A, NRA, Conceal Carry, Family, Kids, Female, Patriotism, Veterans, Military, Police, Western, and Weblink.  Value of 1 was entered if the theme or characteristic was present, zero otherwise.

Software- or Instrument-specific information needed to interpret the data, including software and hardware version numbers:

.csv or .xls files may be opened in spreadsheet applications, such as Excel or Google Sheets, or open source software such as Open Office.

Describe any quality-assurance procedures performed on the data:

Data were coded individually, but simultaneously, frequently working at the same table.  The coders regularly consulted each other on appropriate coding practices and developing a consensus interpretation of characteristics.

--------------------------
DATA-SPECIFIC INFORMATION 
--------------------------

Number of variables:
31

Number of cases/rows: 
top row contains header with variable names
929 rows
928 records

Variable list, defining any abbreviations, units of measure, codes or symbols used:
   
Date: GMT date and time, reported by Twitonomy
Handle: Twitter handle of manufacturer
Name: Name of Twitter owner
Text: Text and emogies contained in tweet
URL: weblink to Twitter post
Platform: original platform used to develop tweet, e.g. Facebook, Instagram, Twitter (by device used), or other source
Type: indicates Tweet, Retweet or Reply
Retweets: Number of retweets
Favorite: Number of times favored
ID: number to indicated order for 20 most recent tweets by manufacturer
RTID: number to indicate order for top 20 tweets retweeted
CCTID: concatenated ID and RTID
Handgun: handgun present
Shotgun: shotgun present
Rifle: rifle present
Attributes: characteristics of firearm described
Protection: tweet indicates or exhibits firearms use for protection
Hunting: tweet indicates or exhibits firearms use for hunting
Recreation: tweet indicates or exhibits firearms use for recreation
2A: tweet indicates or references 2nd Amendment
NRA: tweet indicates or references NRA
Conceal Carry: tweet indicates or references firearms for concealed carry
Family: tweet indicates or exhibits family themes
Kids: tweet indicates or exhibits child use of firearms
Female: tweet includes a woman or quotes a woman
Patriotism: tweet indicates or exhibits patriotic themes (flags, US leaders, American pride)
Veterans: tweet references or exhibits veterans
Military: tweet references or exhibits military themes (soldiers, military use of weapons)
Police: tweet references or exhibits law enforcement themes (police, thin blue line) Western: tweet references or exhibits western or cowboy themes
Weblink: tweet links to website(s) designed to promote sales, such as website(s) of manufacturer or licensed firearms dealers intended to promote gun purchases

Missing data codes: all data were coded
